Re: [Caml-list] Type inference question

[Mark Shinwell <Mark.Shinwell@cl.cam.ac.uk>]

On Tue, Apr 26, 2005 at 03:41:18PM +0200, Andreas Rossberg wrote:
> Mark Shinwell wrote:
> >
> >However, in the Marshal example above, we have
> >
> >	Marshal.from_channel stdin []
> >
> >in the first part of the "let".  In the SML terminology, this is not a
> >"nonexpansive expression"[*] (unlike "fun x -> x").  Therefore, I would
> >have thought that the appearance of such an expression here would
> >prohibit generalisation (in order to prevent possible unsoundness in the
> >presence of mutable state).  This is presumably not the case in OCaml:
> >can someone explain why?
> 
> Recent versions of OCaml employ a slightly relaxed variant of the value 
> restriction that allows generalising of variables in covariant position, 
> because that is always safe.

Ah, yes---I had forgotten about that.  Thanks.

Mark
-- 
Mark Shinwell -- email: Mark.Shinwell@cl.cam.ac.uk
Theory and Semantics Group, University of Cambridge Computer Laboratory